[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [roy@rant-central.com: Re: [arma@mit.edu: Re: Wikipedia & Tor]]



On Thu, Sep 29, 2005 at 03:04:17AM -0700, David Benfell wrote:
> On Thu, 29 Sep 2005 00:38:01 -0400, Nick Mathewson wrote:
> > 
> > This could be achieved, as some people fear, through modifying the
> > core of Tor.  But that isn't the only way to change matters.  As
> > discussed, introducing a separate pseudonymous authentication service
> > (perhaps even an anonymous credential service, if we can find a way to
> > do this without patent infringement) would serve just as well, and
> > require no modifications to the Tor code.  Users who didn't want to
> > use such a service would be no worse off than they are today.  Users
> > who wanted to use Tor and edit Wikipedia at the same time could decide
> > whether the implications of such a service were acceptable to them.
> > 
> Excellent.  You're absolutely right that there would be implications
> to such a service.  That's why it must be kept separate from Tor.  And
> I have no reason to object to what you suggest here.

If you don't include the pseudonym tracking into Tor, the overlap
of people who have both Tor and pseudonym tracking installed will
be very small, and thus have a negligible user base.

At the very least pseudonym tracking should be made an integral
part of Tor installation documentation (just as Privoxy), so that
most users will wind up with installing all of them (assuming,
most Tor operators also have Privoxy installed, of course).

There are two main problems with pseudonym tracking: it must be
implemented as a distributed cryptographic P2P file store with
traffic remixing, so that its operators themselves can't know
what's in the binary blob, and won't be able to modify it.

Another issue is denial of service by e.g. flooding the store
with bogus entries, or using the P2P protocol part to serve
warez, which current Tor tries to prevent. Pseudonym tracking 
would need pseudonym tracking and only track warm bodies, which 
would need a Turing test (CAPTCHa, or similiar). 

As I said, it's not easy to get right. 

-- 
Eugen* Leitl <a href="http://leitl.org";>leitl</a>
______________________________________________________________
ICBM: 48.07100, 11.36820            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE

Attachment: signature.asc
Description: Digital signature